Computers & Education 49 (2007) 10661080

www.elsevier.com/locate/compedu

Surveying instructor and learner attitudes toward e-learning

Shu-Sheng Liaw
a

a,*

, Hsiu-Mei Huang

b,1

, Gwo-Dong Chen

c,2

General Education Center, China Medical University, 91 Shiuesh Road, Taichung, 404, Taiwan, ROC
b
Department of Management Science, National Taichung Institute of Technology, 129, Sec. 3,
Saming Road, Taichung, 404, Taiwan, ROC
c
Department of Computer Science & Information Engineering, National Central University,
No. 300, Jung-da Road, Chung-li, Tao-yuan, Taiwan 320, ROC
Received 23 August 2005; accepted 16 January 2006

Abstract
The trend of using e-learning as a learning and/or teaching tool is now rapidly expanding into education.
Although e-learning environments are popular, there is minimal research on instructors and learners attitudes toward these kinds of learning environments. The purpose of this study is to explore instructors and
learners attitudes toward e-learning usage. Accordingly, 30 instructors and 168 college students are asked
to answer two dierent questionnaires for investigating their perceptions. After statistical analysis, the
results demonstrate that instructors have very positive perceptions toward using e-learning as a teaching
assisted tool. Furthermore, behavioral intention to use e-learning is inuenced by perceived usefulness
and self-ecacy. Regarding to learners attitudes, self-paced, teacher-led, and multimedia instruction are
major factors to aect learners attitudes toward e-learning as an eective learning tool. Based on the ndings, this research proposes guidelines for developing e-learning environments.
2006 Elsevier Ltd. All rights reserved.
Keywords: E-learning; Attitude; 3-TUM; Behavioral intention

Corresponding author. Tel.: +886 4 2205 3366x1927; fax: +886 4 2203 3108.
E-mail addresses: ssliaw@mail.cmu.edu.tw (S.-S. Liaw), hmhuang@ntit.edu.tw (H.-M. Huang), chen@csic.ncu.
edu.tw (G.-D. Chen).
1
Tel.: +886 4 2219 6322.
2
Tel.: +886 3 4226062x35230.
0360-1315/$ - see front matter 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.compedu.2006.01.001

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

1067

1. Introduction
The use of information and Internet technologies as teaching and learning tools is now rapidly
expanding into education. Electronic learning (e-learning) is one of the most popular learning
environments in the information age. Thus, e-learning eorts and experiments currently receive
enormous attention across the globe. Essentially, e-learning, unlike traditional learning, is another
way of teaching and learning. Khan (2000) denes that e-learning encompasses Web-based learning (WBL), Internet-based training (IBT), advanced distributing learning (ADL), and online
learning (OL). Moreover, by another denition, e-learning includes instruction delivered via all
electronic media such as the Internet, intranets, extranets, and hypertext/hypermedia documents
(Govindasamy, 2002). Thus, from e-learning denitions, it acknowledges challenges posed by
diverse of learners and instructors. Indeed, e-learning extends traditional learning paradigms into
new dynamic learning models through computer and Web technologies. However, only a small
amount of e-learning literature assesses both instructors and learners attitudes toward using
e-learning as teaching and learning tools.
Personal attitudes are a major factor to aect individual usage of information technology. In
other words, understanding users attitudes toward e-learning facilitates the creation of appropriate e-learning environments for teaching and learning. Essentially, methods of assessing e-learning
cannot be evaluated using a single linear methodology. In other words, there is a need to build a
multidisciplinary approach to survey individual attitudes toward e-learning (Liaw, 2002; Liaw, in
press; Wang, 2003). The measurement of e-learning must incorporate dierent aspects of user perceptions to form a useful diagnostic instrument (Wang, 2003). Additionally, based on Liaws
(2002) point of view, constructing user attitudes toward computer and Internet technologies
can be divided into three major measurements: aective, cognitive, and behavioral measurements.
The aective measurement (such as perceived enjoyment) and the cognitive measurement (such as
perceived self-ecacy and perceived usefulness) have a positive eect on the behavioral measurement (such as behavioral intention to use e-learning as a teaching or learning tool) (Liaw &
Huang, 2003).
Given the importance of individual attitudes in predicting and improving e-learning usage, this
study raises the need for further research to examine fully the role of instructor and student attitudes in e-learning usage. Thus, the goals of this research are; rst, surveying instructor attitudes
toward e-learning as a teaching assisted tool; second, investigating learner perceptions toward
e-learning as a learning assisted tool; and nally, exploring from the viewpoint of users including
instructors and learners guidelines when deploying e-learning.

2. Literature review
2.1. Developing eective e-learning
In this study, e-learning refers to the use of Internet technologies to deliver a broad array of
solutions that enhance knowledge and performance. Rosenberg (2001) states that e-learning is
based on the following three fundamental criteria: rst, e-learning is networked, making it capable
of instant updating, storage/retrieval, distribution, and sharing of instruction or information;

1068

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

second, it is delivered to the end-user via a computer using standard Internet technology; and
third, it focuses on the broadest view of learning that exceeds beyond the traditional paradigms
of training. Thus, in e-learning, learning activities center around learner autonomy and interactive
learning actions; in addition, learning instruction is based on multiple media and ill-structured
formats. Furthermore, it also oers cooperative learning opportunities between instructors and
learners and/or among learners. Based on Rosenbergs approach, e-learning oers more opportunities for improving problem solving capabilities, enhancing high order thinking skills, and
achieving learning eectiveness (Chen, Lee, & Chen, 2005; Liaw, 2004).
For designing eective e-learning environments, Liaw (2004) suggested three considerations:
learner characteristics, instructional structure, and interaction. In developing e-learning, it is
necessary to understand targeted population. First, learner characteristics, such as attitudes,
motivation, belief, and condence need to be identied (Passerini & Granger, 2000). Essentially,
e-learning signies autonomous learning environments. In other words, users have more opportunities for self-directed learning in e-learning environments. As for instructional structure, multimedia instruction enables learners to develop complex cognitive skills, such as understanding
important elements of conceptual complexity, ability to use acquired concepts for reasoning
and inference, and competence to apply conceptual knowledge to novel situations with exibility
(Spiro, Feltovich, Jacobson, & Coulson, 1995). Finally, e-learning environments oer group interaction, such as learners to learners, or learners to instructors. Group interaction is a kind of cooperative learning that helps learners to make progress through their zone of proximal development
by the activities in which they engage (Vygotsky, 1978). When learners increase their interaction
with instructors and learners, they in turn raise their chances of building their own knowledge
because much of learning inevitably takes place within a social context, and the process includes
the mutual construction of understanding (Bruner, 1971). Thus, based on fundamental e-learning
criteria, these are three considerations in designing e-learning environments: autonomous learning, multimedia environments, and teacher-assisted learning. Fig. 1 presents considerations when
facilitating e-learning.
2.2. Attitudes toward e-learning
Many institutions of higher education are resorting to e-learning for authentic learning and
enhancing learning performance, while other schools are hopping onto the bandwagon simply

Autonomous

Multimedia

Instructor-led

learning

environments

learning

Creating effective e-learning environments

Fig. 1. Considerations for developing eective e-learning.

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

The first tier: the tier of

The second tier:

The third tier:

individual experience and

the affective and

behavioral

system quality

cognitive tier

intention tier

1069

Fig. 2. The three-tier use model (3-TUM).

because they do not want to be left behind (Govindasamy, 2002). When instructors exhibit more
positive attitudes toward e-learning, then they have more behavioral intentions to use it (Liaw, in
press). Indeed, no matter how advanced or capable the technology is, its eective implementation
depends upon users having a positive attitude toward it. Thus, as individuals attitudes on e-learning
become more positive, they will have greater behavioral intention to use it.
Although the concept of attitude towards computers has gained recognition as a critical determinant in the use and acceptance of computer technology, there is no single, universally accepted
denition of computer attitude construct (Liaw, 2002; Smith, Caputi, & Rawstorne, 2000). Previous research (Triandis, 1971) suggested that attitude consists of aective, cognitive, and behavioral components. The aective component is the emotion or feeling which includes statements
of likes or dislikes toward certain objects. The cognitive component refers to statements of beliefs.
And the behavioral component is what an individual actually does or intends to do (Liaw, 2002).
3-TUM (three-tier technology use model) is a conceptual approach for investigating user perceptions toward information and Internet technologies (Liaw, in press). The original concept of 3TUM is derived from TAM (Technology Acceptance Model). TAM is a popular approach for
surveying user attitudes of information technologies (Davis, Bagozzi, & Warsaw, 1989). TAM
suggests that two specic behavioral beliefs, perceived ease of use (EOU) and perceived usefulness
(U), determine an individuals behavioral intention to use technologies. Based on the 3-TUM
(Fig. 2), individual attitudes toward information technology form three dierent tiers: the tier
of individual experience and system quality, the aective and cognitive tier, and the behavioral
intention tier. The tier of individual experience and system quality sets out to evaluate how individual experience and system quality inuence individual aective and cognitive components. The
aective and cognitive tier investigates how aective and cognitive components change individual
behavioral intentions. In the behavioral intention tier, the 3-TUM predict individual behavioral
intention to use technology for a particular purpose (for instance search engines as a learning
assisted tool or computers as a job assisted tool) (Liaw, in press).
3. Research hypotheses
3.1. Hypotheses of instructor attitudes
Based on the conceptual models of 3-TUM, this research integrates these two approaches to
understand instructors attitudes toward e-learning environments.
From the concepts of 3-TUM models, system quality is a key factor to inuence individual
aective and cognitive components. The aective and cognitive attitudes investigate how aective
and cognitive attitudes change individual behavioral intentions. DeLone and McLean (1992)
suggest that system quality measures information quality and system quality, both crucial

1070

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

constructs related to the success of information systems. Since e-learning environments are multimedia systems, information quality can be viewed as various information formats. Additionally,
user attitudes toward e-learning environments, such as perceived satisfaction or ease of use are
functions of system quality. Therefore, this study proposes multimedia environments and system
satisfaction as variables that express the system quality of e-learning environments.
Based on 3-TUM, this study proposes the following hypotheses.
H1A: The quality of e-learning is positively related to instructors perceived enjoyment toward it.
H1B: The quality of e-learning is positively related to instructors perceived usefulness toward it.
H1C: The quality of e-learning is positively related to instructors perceived self-ecacy toward it.
Furthermore, based on 3-TUM approach, perceived enjoyment, and perceived usefulness are
positively related to individual behavioral intention to use information systems. Relevant studies
(Liaw & Huang, 2003; Moon & Kim, 2001; Vankatesh, 1999) also support these results. Moreover, Liaw (in press) nds that perceived self-ecacy is a major factor to inuence individual
behavioral intention to use information systems. Therefore, this study leads to three hypotheses:
H1D: The instructors perceived enjoyment toward e-learning is positively related to their intentions to use it.
H1E: The instructors perceived usefulness toward e-learning is positively related to their intentions to use it.
H1F: The instructors perceived self-ecacy toward e-learning is positively related to their intentions to use it.

3.2. Hypotheses of learner attitudes

Liaw (2004) suggested three considerations when developing eective e-learning environments:
learner characteristics, instructional structure, and interaction. E-learning provides a user more
opportunities to be an active and self-regulatory learner. For example, in an e-learning environment, learners can control the learning time and procedures by themselves. Thus, based on learner
perspectives, e-learning is a learner autonomy environment. In addition, e-learning environments
provide various assisted functions, such as teacher-made online instruction, online conference,
online help and suggestions, online examination, and online monitoring. All these functions oer
opportunities for teachers to be assisted tutors. Furthermore, e-learning environments oer both
multimedia ill-structured and well-structured instructions. As e-learning environments usually
provide collaborative, interactive, network systems, and multimedia contents, they oer a learning
environment to improve learners problem-solving capabilities and thinking skills. Thus, based on
e-learning fundamental criteria, three guidelines should be noted: self-paced, instructor-led, and
multimedia for building eective learning environments.
Therefore, this study proposes three hypotheses:
H2A: Self-paced learning is a predictor to learners attitudes toward e-learning as an eective
learning tool.

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

1071

H2B: Instructor-led learning is a predictor to learners attitudes toward e-learning as an eective

learning tool.
H2C: Multimedia instruction is a predictor to learners attitudes toward e-learning as an eective
learning tool.
4. Research methodology
4.1. Understanding instructor attitudes
4.1.1. Participants
The study was conducted in a university with a sample of 50 instructors. These 50 instructors
attended a seminar on How to Use E-Learning for Teaching Purposes. All subjects are asked to
answer a questionnaire after the 4-h seminar. The questionnaire with a cover letter was distributed
to subjects from the researchers. All respondents were asked to complete the survey and their feedback was guaranteed condentiality. Thirty-ve instructors answered the questionnaire. Five
missing responses were eliminated and a total of 30 responses were collected.

4.1.2. Instruments
The data for this study was gathered by means of a questionnaire. The questionnaire included
three major components: (a) demographic information, (b) computer and Internet experience, and
(c) attitudes toward e-learning. The questionnaire is described as below.
Demographic information: The demographic component covered gender, and the eld of
teaching.
Computer and Internet experience: In this component, participants were asked to indicate whether
they had experience using operating systems, the Internet, word processing packages, PowerPoint,
computers as a teaching assisted tool, and experience using e-learning. These 6 questions are all 7point likert scales (from 1 which means no experience to 7 which means well experienced).
Attitudes toward e-learning: Participants were asked to indicate their attitudes toward e-learning. These 19 questions were all 7-point likert scales (from 1 which means strongly disagree to 7
which means strongly agree).

4.2. Understanding learner attitudes

4.2.1. Participants
This study conducts another questionnaire survey for understanding learner attitudes. The survey was distributed to 168 college students who took either Introduction to Computer Science
or Introduction to Computer Network courses. After using the e-learning environment for six
weeks, all participants were asked to complete a questionnaire that included demographic information and three dierent components (demographic information, computer and Internet experience, and attitudes toward e-learning). The questionnaires, including a cover letter, were
distributed to participants during class. All subjects were asked to respond to the questionnaire
and their responses were guaranteed condentiality.

1072

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

4.2.2. Instruments
The data for this study was gathered by means of a paper-and-pencil survey. The questionnaire included three major components: (a) demographic information, (b) computer and Internet experience, and (c) attitudes toward e-learning. The following shows the content of the
questionnaire.
Demographic information: The demographic component covered gender and the eld of study.
Computer and Internet experience: In this component, participants were asked to indicate
whether they had experience using Web browsers, e-mail, word processing packages, as well as
coding Web pages. These 4 questions are all 7-point likert scales (from 1 which means no experience to 7 which means well experienced).
Attitudes toward e-learning: Participants were asked to indicate their attitudes toward e-learning. These 15 questions all adopted 7-point likert scales (from 1 which means strongly disagree
to 7 which means strongly agree).

5. Results
5.1. Results of instructors attitudes
Regarding instructors experience in using computers or Internet technology for teaching, only
one instructor does not have any experience while the other 29 instructors all have experience.
Descriptive statistics (means (M) and standard deviations (SD)) of computer use and experience
were shown in Table 1. The alpha reliability of instructor attitudes toward e-learning (Table 2)
was to be highly accepted (a = 0.95). The high alpha reliability gives a support for questionnaire
content reliability.
The Pearson correlation coecients among the variables are presented in Table 3. The bivariate relationships indicated that many of the variables signicantly correlated with each other,
but are all less than 0.80.
Concerning analytic strategy for assessing the predictive model, multiple regression analysis is
an appropriate multivariate analytical methodology. Essentially, multicollinearity can be controlled in two ways: (1) correlation between independent variables should all be less than 0.8
(Emory & Cooper, 1991); (2) variance ination factors (VIF) should be less than 10 (Neter &
Kutner, 1990). In this study, multicollinearity was ruled out because the correlation between independent variables were all less than 0.8 and the VIFs were all less than 10. Based on multiple
regression analysis, the scatter plots of the standardized residuals by the standardized predicted
scores were also examined to verify the assumption of linearity.
The results of stepwise multiple regressions for the path associated with the variables were presented in Table 4. For understanding H1A, H1B, and H1C, three regression analyses were performed to check the eects of predicted variables (perceived system satisfaction and multimedia
instruction) on perceived self-ecacy, perceived enjoyment, and perceived usefulness toward
e-learning environments. The results showed the independent variables of perceived system satisfaction could predict perceived self-ecacy (F(1, 28) = 10.58, p < 0.01, R2 = 0.27), and perceived
usefulness (F(1, 28) = 68.74, p < 0.01, R2 = 0.71). In addition, the independent variables of multimedia instruction could predict perceived enjoyment (F(1, 28) = 13.50, p < 0.01, R2 = 0.33). For

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

1073

Table 1
Descriptive statistics of instructor computer use and experience (from 1 which means no experience to 7 which means
well experienced)
Variables
Experience
Experience
Experience
Experience
Experience
Experience

using
using
using
using
using
using

operating systems
the Internet
word processing packages
PowerPoint
computers as a teaching assisted tool
e-learning

SD

5.33
5.47
5.67
5.40
4.73
4.37

1.42
1.14
1.15
1.28
1.91
2.01

Table 2
The item-total correlations of instructor attitudes (from 1 which means strongly disagree to 7 which means strongly
agree)
Items

SD

r*

Perceived self-ecacy
I feel condent making online instruction
I feel condent using the Internet
I feel condent using e-learning environments

5.20
5.13
5.40
5.07

1.19
1.46
1.22
1.28

0.73
0.63
0.51

Perceived enjoyment
I enjoy using computers as a teaching assisted tool
I enjoy using e-learning environment for teaching purpose
I enjoy using online instruction for teaching

5.69
5.63
5.66
5.77

1.04
1.27
0.96
1.36

0.49
0.67
0.63

Perceived usefulness
I believe using e-learning environments is helpful for learning
I believe using e-learning environments is helpful for teaching
I believe using online instruction is useful for teaching

5.89
6.13
5.77
5.77

0.98
1.07
1.07
1.04

0.76
0.56
0.73

Behavioral intention to use e-learning

I intend to use e-learning to assist my teaching
I intend to use online instruction to assist my teaching
I intend to use the Internet to assist my teaching

5.71
5.53
5.53
6.07

1.00
1.17
1.06
1.14

0.81
0.77
0.74

Perceived system satisfaction

I am satised with using e-learning environments
I am satised with using MS-Word, MS-PowerPoint les as multimedia instruction
I am satised with using online instruction

5.87
5.90
5.90
5.80

0.91
0.99
1.06
0.92

0.69
0.85
0.82

Multimedia instruction
I like to use voice media instruction
I like to use image media instruction
I like to use animation media instruction
I like to use colorful text media instruction

5.96
5.77
5.93
6.03
6.10

0.93
1.10
1.01
0.96
0.88

0.80
0.81
0.80
0.83

r*: Corrected item-total correlation.

examining H1D, H1E, and H1F, a regression analysis was performed to check the eects of
predicted variables (perceived self-ecacy, perceived enjoyment, and perceived usefulness) on
behavioral intention to use e-learning environments. The results showed the independent variables

1074

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

Table 3
Correlation analysis of instructor attitudes
Variables
1.
2.
3.
4.
5.
6.
**

Perceived self-ecacy
Perceived enjoyment
Perceived usefulness
Behavioral intention to use e-learning
Perceived system satisfaction
Multimedia instruction

0.64

3
**

5
**

0.32
0.31

6
**

0.68
0.58**
0.75**

0.52
0.54**
0.78**
0.76**

0.50**
0.57**
0.73**
0.79**
0.78**

Correlations are signicant at the p < 0.01 (2-tailed).

Table 4
Regression results of instructor attitudes
Dependent variable

Independent variables

R2 change

Perceived self-ecacy
Perceived usefulness
Perceived enjoyment
Behavioral intention to use e-learning

Perceived system satisfaction

Perceived system satisfaction
Multimedia instruction
Perceived usefulness
Perceived self-ecacy

0.68
0.91
0.63
0.60
0.41

0.27
0.71
0.33
0.56
0.21

0.003
0.000
0.001
0.000
0.000

of perceived usefulness and perceived self-ecacy could predict instructors behavioral intention
to use e-learning (F(2, 27) = 46.22, p < 0.0005, R2 = 0.77), and perceived usefulness was the greatest contributor (R2 = 0.56).
5.2. Results of learners attitudes
According to learners experience in using e-learning environments, 73 students have not taken
any e-learning course, 80 students have taken one e-learning course, 9 students two e-learning
courses, and 6 students three or more e-learning courses. Descriptive statistics (means (M) and
standard deviations (SD)) of computer use and experience were shown in Table 5. The alpha reliability of learner attitudes toward e-learning (Table 6) was to be highly accepted (a = 0.92). The
high alpha reliability gives a support for questionnaire content reliability.
The Pearson correlation coecients among the variables are presented in Table 7. The bivariate relationships indicated that many of the variables signicantly correlated with each other.
Table 5
Descriptive statistics of learner attitudes (from 1 which means no experience to 7 which means well experienced)
Variables
Experience
Experience
Experience
Experience

using Web browsers

using e-mail
using word processing packages
coding Web pages

SD

4.92
5.46
4.89
3.11

1.70
1.41
1.48
1.50

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

1075

Table 6
Means, standard deviations and item-total correlations (from 1 which means strongly disagree to 7 which means
strongly agree)
Items

SD

r*

E-learning as a self-paced learning environment

I can learn actively in the e-learning environment
I have more opportunities to create my own knowledge in the e-learning environment
The hypertext online instruction can enhance my learning motivation
I can discuss actively with others in the e-learning environment
I can read the online instruction actively
I can nd information actively in the e-learning environment

5.16
4.77
5.62
4.70
5.35
5.15
5.34

1.37
1.43
1.24
1.39
1.26
1.44
1.43

0.57
0.75
0.66
0.49
0.68
0.68

E-learning as an eective learning environment

The e-learning environment improves my thinking skills
The e-learning environment enhances my problem-solving skills
The e-learning environment provides various aspects to solve problems

5.31
5.04
5.42
5.46

1.26
1.39
1.23
1.16

0.69
0.75
0.62

E-learning as a multimedia instruction environment

I like colorful pictures in online instruction
I like learning videos in online instruction
I like the animated online instruction

5.93
6.04
5.90
5.86

1.15
1.09
1.21
1.15

0.62
0.66
0.60

Teachers as an instructor-led learning environment

I like the instructors help and suggestions in the e-learning environment
I like the instructors voice and image in the e-learning environment
I like the instructors online multimedia instruction in the e-learning environment

4.68
4.89
4.51
4.65

1.54
1.48
1.60
1.54

0.60
0.60
0.60

r*: Corrected item-total correlation.

Concerning analytic strategy for assessing H2A, H2B, H2C, multiple regression analysis is an
appropriate multivariate analytical methodology for empirically examining sets of relationships
in the form of linear causal models. The results of stepwise multiple regressions for the path associated with the variables were presented in Table 8. The regression analysis was performed to
check the eects of predicted variables (e-learning as a self-paced learning environment, e-learning
as a multimedia instruction environment, and e-learning as an instructor-led learning environment) on e-learning as an eective learning environment. The results showed all three independent
variables are predictors (F(3, 163) = 67.21, p < 0.0005, R2 = 0.55). These three predictors have
55% contribution.

Table 7
Correlation analysis of learner attitudes
Variables
1.
2.
3.
4.
**

E-learning
E-learning
E-learning
E-learning

2
as
as
as
as

a self-paced learning environment

an eective learning environment
a multimedia instruction environment
an instructor-led learning environment

Correlations are signicant at the p < 0.01 (2-tailed).

0.66

3
**

4
**

0.60
0.62**

0.54**
0.57**
0.54**

1076

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

Table 8
Regression results of learner attitudes
Dependent variable

Independent variables

R2 change

E-learning as an eective
learning environment

E-learning as a self-paced learning

environment
E-learning as a multimedia
instruction environment
E-learning as an instructor-led
learning environment

0.38

0.44

0.000

0.28

0.08

0.000

0.23

0.03

0.001

6. Discussions
The results of this research conrm that all Hypotheses are true, from H1A through H2C.
Based on descriptive statistical data of Table 1, instructors have adequate computer experience
(such as operating system and word processing experience). Furthermore, they are also well experienced in using of the Internet. According to teaching technology experience (such as experience
using PowerPoint, computers as a teaching assisted tool, and experience using e-learning), instructors have high competency in using those tools. These results support that instructors have
remarkable computer and e-learning related skills for teaching.
Additionally, from Table 2, instructors have highly positive attitudes toward e-learning that
included perceived self-ecacy, enjoyment, usefulness, and behavioral intention of use. As for
the quality of e-learning environment, instructors believe both system satisfaction and multimedia
instruction are vital factors. Thus, from Table 2, instructors display highly favorable attitudes
toward e-learning environments as a useful teaching assisted tool, supporting their intention to
use them for teaching in the future. Table 3 presents that these 6 factors have high correlations
with each other. These results indicate that these 6 factors could be investigated at the same time
when understanding users attitudes toward e-learning.
As shown in Table 4, perceived system satisfaction is a crucial factor to inuence instructors
perceived self-ecacy and perceived usefulness toward e-learning. Multimedia instruction is a
critical predictor for their perceived enjoyment. Furthermore, when understanding instructors
behavioral intention to use e-learning environments, perceived usefulness is the greatest contributor (56%) and perceived self-ecacy has a 21% contribution. From instructors viewpoints, system satisfaction can positively aect their attitudes toward e-learning as a useful teaching assisted
tool. And multimedia instruction can signicantly aect their enjoyment toward e-learning.
On the other hand, Table 5 shows that learners have abundant computer related experience.
Students have more experience using browsers and e-mail than word processing. This rearms
that todays learners have adequate Internet related experience. From Table 6, students feel that
e-learning environments can be autonomous and rich in multimedia. Additionally, they expect
teachers to help them when they use e-learning environments for assisted learning. Moreover, students believe that e-learning environments are an ecient learning tool. Therefore, they have positive attitudes toward e-learning environments for learning purposes as shown in Table 6. Table 7
presents that 4 factors (self-paced, multimedia instruction, instructor-led, and eective learning)
have high correlations with each other. These results indicate that these factors could be investigated at the same time when understanding learner attitudes toward e-learning.

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

1077

Table 8 shows learners attitudes toward e-learning as an ecient learning tool can be predicted
positively by three factors (e-learning as a self-paced learning environment, e-learning as a form of
multimedia instruction, and e-learning as an instructor-led learning environment). When building
an e-learning environment, these three factors should be taken into consideration in order to
create an ecient learning environment.

7. Conclusions
Modern day learning environments are characterized by their place and time independence,
their integrated presentation and communication facilities, and their opportunities for re-use of
learning technologies in the form of learning objects. Many researchers claim that technology
push will enhance the quality of education; In fact, Clark (1994) argues that the question of
whether media or technology will ever inuence learning remains open to debate. A well-defendable
viewpoint lies not in the media or technology used because only positive attitudes toward that
media or technology can improve the quality of learning or teaching. Thus, understanding users
attitudes toward learning technology, including instructors and learners attitudes, enables us to
make learning more eective, ecient, and appealing.
When applying a learning tool or system for learners, it is necessary to investigate both teachers and learners attitudes toward that tool or system. Essentially, understanding their perceptions toward learning environments is a crucial issue for enhancing teaching performance and
learning eects. The results of this study conrm that instructors are willing to use e-learning environments to aid their teaching activities. Learners also respond favorably to e-learning environments for complementing to their learning activities. The present study helps us understand
instructor intentions and learner attitudes toward e-learning. Specically, user perceptions toward
e-learning could include aective, cognitive, behavioral, and social components. All in all, e-learning
environments could be developed with these guidelines in mind: multimedia instruction, autonomous learning, instructor-led interaction, and improving learning eectiveness.
7.1. Understanding instructors intentions to use e-learning
The results of this research provide evidence that instructors have highly behavioral intentions
to use e-learning (mean is 5.71). In other words, e-learning could be a teaching assisted tool. From
the ndings, perceived e-learning satisfaction is a key factor to aect instructors cognitive perceptions, such as perceived self-ecacy and perceived usefulness of e-learning. Table 4 also shows
that multimedia instruction is a main factor to inuence instructors aective perception. Regarding instructors behavioral intention to use e-learning, perceived usefulness and self-ecacy are
two major predictors. Thus, the results support that 3-TUM is an appropriate research model
for investigating instructors attitudes of e-learning usage.
Among the various theoretical models developed to examine users intentions of using
computer and communication technology, perceived usefulness is a key to inuence behavioral
intentions (Gefen & Straub, 1997; Liaw & Huang, 2003; Moon & Kim, 2001; Szajna, 1996; Taylor
& Todd, 1995; Vankatesh & Davis, 1996; Vankatesh, 1999). According to Bandura (1997),
self-ecacy is the belief in ones capabilities to organize and execute the courses of action

1078

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

required to produce given attainments (p. 3). Some researchers emphasize that self-ecacy
beliefs should be assessed in such a way that the beliefs correspond to the targeted performance
and domain of interest. Compeau and Higgins (1995a) state that when individuals have higher
self-ecacy toward information technology, they also feel that information technology is more
useful. Additionally, Compeau and Higgins (1995b) and Compeau et al. (1999) indicate that when
individuals have higher self-ecacy toward information technology, they also intend to use information technology more. Therefore, the ndings of this research support previous studies.
7.2. Realizing learner attitudes toward e-learning
One of the critical factors using e-learning is to improve learning performance. The results of this
study oer a view that eective e-learning environments are aected by learner self-paced learning,
multimedia instruction, and instructor-led learning. It means that an eective learning environment, including problem-solving capabilities and high order thinking skills, is inuenced by learner
autonomy, multimedia content, and teacher helps. Although self-paced learning, multimedia
instruction, and instructor-led learning are all major predictors, learners have the highest positive
attitudes toward multimedia instruction (mean is 5.93) and moderately positive attitudes toward
instructor-led learning (mean is 4.68). From the results, it supports that most learners prefer multimedia instruction and many of them believe that interaction with others is necessary. Pallo and
Pratt (1999) state that the main features of the learning process include interactions among students, and interactions between students and instructors for collaboration in learning. Therefore,
e-learning should provide useful tools and instructors should utilize creative teaching approaches
that can facilitate the interactions and collaboration in the learning process, which may nally
increase eective learning performance, such as problem-solving and high order thinking.
7.3. Perceptions toward e-learning including aective, cognitive, behavioral, and social components
Much research on using learning technologies in education involves some measure of learners
attitudes by means of a questionnaire or an interview after they have used a particular piece of
technology or software. In such questionnaires, researchers often ask users about their experience.
Based on questionnaire investigation, users responses serve as critical information to understand
more about users attitudes toward that technology or software. Liaw (2002) claims that constructs of user attitudes toward computer and Internet technologies should be divided into three
major measurements: aective, cognitive, and behavioral measurements. Additionally, Jones and
Issro (2005) argue the importance of considering both aective and social components when trying to understand user attitudes. Thus, user attitudes toward e-learning could be studied from various perspectives, such as aective, cognitive, behavioral, and social components.
This research provides aective, cognitive, behavioral, and social perspectives to understand
instructors and learners attitudes toward e-learning. Attention to these four aspects would serve
as an ideal beginning to further our comprehension of the role of attitude factors when using
learning technologies. From the results of social perspectives, instructors leadership or assistance
is a critical factor to aect learners attitudes. In other words, interaction between instructors and
students is a success factor for enhancing e-learning eects. Based on activity theory, learning
activities include individual and social levels at the same time (Kuutti, 1997).

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

Highly

Vivid

Enhancing

Improving

autonomous

multimedia

teacher-learner

learning

learning

instruction

communication

effectiveness

1079

Developing effective e-learning environments

Fig. 3. Guidelines of developing eective e-learning.

7.4. Guidelines of developing e-learning environments

This research oers four factors that could be considered when developing e-learning environments: multimedia instruction, autonomous learning, instructor-led interaction, and improving
learning eectiveness. Based on the results of investigating instructors and learners attitudes, four
guidelines should be considered when facilitating e-learning: vivid multimedia learning instruction,
highly autonomous learning environments, enhancing instructors and learners asynchronous or
synchronous communications, as well as improving learning eectiveness (including learners problem solving or high order thinking skills presented in Table 6). Fig. 3 shows the guidelines of developing e-learning environments.
In summary, e-learning is the new wave in learning strategy. Through innovative use of modern
technology, e-learning not only revolutionizes education and makes it more accessible, it also
brings formidable challenges for instructors and learners. Thus, surveying instructors and students attitudes toward e-learning is a critical issue in learning and training. This research assesses
instructors attitudes toward e-learning environments as a teaching assisted tool. Moreover, this
study explores learners perceptions toward e-learning as an eective learning tool. Future
research may explore methods to increase learner autonomy, and interactions among instructors
and students for improving learning performance. Besides, how to design vivid multimedia
e-learning contents for improving learners performance is also a potential research direction.

Acknowledgement
This study was partially supported by the National Science Council, Taiwan, ROC, Project
Nos. NSC93-2524-S-039-001, NSC94-2520-S-039-001 and NSC94-2524-S-008-002.

References
Bandura, A. (1997). Self-ecacy: The exercise of control, USA.
Bruner, J. S. (1971). The relevance of education. Cambridge, MA: Harvard University Press.
Chen, C. M., Lee, H. M., & Chen, Y. H. (2005). Personalized e-learning system using Item Response Theory.
Computers & Education, 44, 237255.
Clark, R. E. (1994). Media will never inuence learning. Educational Technology Research and Development, 42(2),
2129.

1080

S.-S. Liaw et al. / Computers & Education 49 (2007) 10661080

Compeau, D. R., & Higgins, C. A. (1995a). Application of social cognitive theory to training for computer skills.
Information Systems Research, 6(2), 118143.
Compeau, D. R., & Higgins, C. A. (1995b). Computer self-ecacy: development of a measure initial test. MIS
Quarterly, 19(2), 189211.
Compeau, D. R., Higgins, C. A., & Hu, S. (1999). Social cognitive theory and individual reactions to computing
technology: a longitudinal study. MIS Quarterly, 23(2), 145158.
Davis, F. D., Bagozzi, R. P., & Warsaw, P. R. (1989). User acceptance of computer technology: a comparison of two
theoretical models. Management Science, 35(8), 9831003.
DeLone, W., & McLean, E. (1992). Information systems success: the quest for the dependent variable. Information
Systems Research, 3(1), 6095.
Emory, C. W., & Cooper, D. R. (1991). Business research method (4th ed.). Homewood, IL: Irwin.
Gefen, D., & Straub, D. W. (1997). Gender dierences in the perception and use of e-mail: an extension to the
technology acceptance model. MIS Quarterly, 21(4), 389400.
Govindasamy, T. (2002). Successful implementation of e-learning pedagogical considerations. The Internet and Higher
Education, 4, 287299.
Jones, A., & Issro, K. (2005). Learning technologies: aective and social issues in computer-supported collaborative
learning. Computers & Education, 44, 395408.
Khan, B. H. (2000). A framework for web-based learning. Englewood Clis, NJ: Educational Technology Publications.
Kuutti, K. (1997). Activity theory as a potential framework for humancomputer interaction research. In B. A. Nardi
(Ed.), Context and consciousness: Activity theory and humancomputer interaction. Cambridge, MA: MIT Press.
Liaw, S. S. (2002). An Internet survey for perceptions of computer and World Wide Web: relationship, prediction, and
dierence. Computers in Human Behavior, 18(1), 1735.
Liaw, S. S. (2004). Considerations for developing constructivist Web-based learning. International Journal of
Instructional Media, 31(3), 309321.
Liaw, S. S. (in press). Understanding computers and the Internet as a work assisted tool. Computers in Human Behavior.
Liaw, S. S., & Huang, H. M. (2003). An investigation of users attitudes toward search engines as an information
retrieval tool. Computers in Human Behavior, 19(6), 751765.
Moon, J. W., & Kim, Y. G. (2001). Extending the TAM for a World-Wide-Web context. Information & Management,
38, 217230.
Neter, J., & Kutner, L. (1990). Applied linear statistical models. Homewood, IL: Irwin.
Passerini, K., & Granger, M. J. (2000). A development model for distance learning using the Internet. Computers &
Education, 34, 115.
Pallo, R. M., & Pratt, K. (1999). Building learning communities in cyberspace: Eective strategies for the online
classroom. San Francisco, CA: Jossey-Bass.
Rosenberg, M. J. (2001). E-learning, strategies for delivering knowledge in the digital age. New York: McGraw-Hill.
Smith, B., Caputi, P., & Rawstorne, P. (2000). Dierentiating computer experience and attitudes toward computers: an
empirical investigation. Computers in Human Behavior, 16, 5981.
Spiro, R. J., Feltovich, P. J., Jacobson, M. J., & Coulson, D. K. (1995). Cognitive exibility, constructivism, and
hypertext: random access instruction for advanced knowledge acquisition in ill-structured domains. Educational
Technology, 31(5), 2433.
Szajna, B. (1996). Empirical evaluation of the revised technology acceptance model. Management & Science, 42(1), 8592.
Taylor, S., & Todd, P. A. (1995). Understanding information technology usage: a test of competing models. Information
Systems Research, 6(2), 144176.
Triandis, H. C. (1971). Attitude and attitude change. New York: Wiley.
Vankatesh, V., & Davis, F. D. (1996). A model of the antecedents of perceived ease of use: development and test.
Decision Sciences, 27(3), 451481.
Vankatesh, V. (1999). Creation of favorable user perceptions: exploring the role of intrinsic motivation. MIS Quarterly,
23(2), 239260.
Vygotsky, I. S. (1978). Mind in society. Cambridge, MA: Harvard University Press.
Wang, Y. S. (2003). Assessment of learner satisfaction with asynchronous electronic learning systems. Information &
Management, 41, 7586.